Introduction

Many cancers and many treatments for cancer are painful. Pain occurs in 50% of patients with cancer, and approximately 10 to 20% experience severe pain [1, 2]. Nearly one-third of cancer survivors experience chronic pain, defined as experiencing pain on most days for 90 days or longer [3]. Pain is frequently treated with prescription opioids. Although opioids are the most effective treatment for moderate to severe pain and are generally safe with short-term use, long-term opioid therapy carries numerous mental and physical health risks, including increased risk for new-onset depression and worsening of existing depression [4,5,6,7,8].

Independent of long-term opioid therapy, depression is common in adult patients with cancer. Depression is prevalent in 20–30% of cancer patients [9] and nearly 25% of patients in palliative settings [10]. Furthermore, depression is undertreated in oncology settings. In a large UK sample, 73% of cancer outpatients with depression were not receiving potentially effective depression treatments [11]. Depression is a risk factor for long-term opioid therapy and a key contributor to adverse opioid-related outcomes such as dependence and overdose [12,13,14,15]. In addition, long-term opioid therapy in non-cancer pain has been shown to increase risk for new-onset and worsening depression [4,5,6,7,8, 16,17,18]. This bi-directional association is consistent with the concept of hyperkatifeia.

Hyperkatifeia

Hyperkatifeia is a term first reported by Shurman and Koob [19, 20••] and derives from the Greek word, katifeia, which means dejection or low emotional state. Hyperkatifeia is a negative emotional state involving malaise, irritability, unease, dysphoria (general dissatisfaction with life), alexithymia (inability to describe emotions), and anxiety [19, 20••]. Although not a diagnosis, hyperkatifeia’s symptoms overlap with depression, anhedonia (inability to feel pleasure), and dysthymia (persistent mild depression).

It has been postulated [19, 20••] that hyperkatifeia underlies the negative reinforcement behind drug seeking. In the case of prescription opioids, negative reinforcement in the form of pain and dysphoria during withdrawal motivates opioid use to return to hedonic homeostasis and euthymia. As proposed in Figure 1, with long-term opioid therapy, the ability to experience natural rewards is reduced, and the threshold for feeling normal, or not distressed or depressed, increases. Thus, higher opioid doses are required to maintain homeostasis. Over time, opioids no longer reduce pain and may lead to hyperalgesia and dysphoria. Consequently, negative reinforcement becomes the driver of opioid seeking, as opposed to positive reinforcement from pain relief and opioid-induced euphoria [19, 20••].

Fig. 1
figure 1

Provisional pathway from pain to increasing opioid use, depression, and morbidity and mortality.

Full size image

Hyperkatifeia in long-term opioid therapy is consistent with epidemiological research demonstrating that psychiatric disorders are associated with an increased risk for long-term opioid use and opioid misuse, abuse, and use disorder [12, 14, 21]. Longer-term use compared to short-term use increases risk for new onset depression, depression recurrence, and worsening depression [4,5,6, 17]. In non-cancer settings, patients who experienced an increase in opioid dose over 2 years were also more likely to have an increase in depression symptoms [22]. Rapid dose escalation vs. stable dose has also been linked to increased risk for new-onset depression [16], as has pre- and post-operative opioid use [7]. A Mendelian randomization study supports a causal association between genetic liability for greater prescription opioid use and risk of major depressive disorder [8]. While hyperkatifeia has not been examined in cancer populations, this model provides an initial framework to understand patterns of opioid use and depression symptoms in cancer pain populations.

Consequences of depression in cancer pain

While studies of long-term opioid use, hyperkatifeia, and depression risk have not been conducted in patients with cancer pain, we can extrapolate findings to this patient population. For patients with cancer, long-term opioid use may lead to depression that in turn complicates recovery and increases risk for mortality. Cancer pain is associated with a 3.5 times greater risk for feeling depressed and 2.5 times risk for feeling anxious [1, 3]. Pain interference continued to predict depression in a sample of largely male patients 1 year after cancer diagnosis [23]. Depression is associated with more than twice the risk of hospitalization for opioid poisoning and opioid abuse or dependence among patients with cancer pain. In a cohort with a mean survival time of 318 days, those with more severe depression symptoms had a 75% increased risk of death [24]. Risk of suicide and suicide mortality is particularly elevated in head and neck cancer patients and might be driven by pain and depression [25, 26]. In patients with advanced cancer, both depression and pain are associated with worse quality of life, with depression identified as the strongest single predictor among a range of factors, including systemic inflammation, physical performance, symptom burden, and comorbidities [27]. A more encouraging implication of this association is that effective treatment of depression also improves cancer pain management. In a study of 274 cancer patients, greater improvement in depression predicted pain improvement over a 12-month period (OR 1.84) [28].

Detecting depression in cancer pain

Pain can mask depression [29], and patients may blame pain for their depression and, as in the case of hyperkatifeia, mistakenly interpret opioid-induced euphoria as an antidepressant effect. A critical first step in mitigating risk of opioid-related emotional disturbance is to screen for depression, dysthymia, and anhedonia, not only before initiating opioids but repeatedly during opioid therapy.

Most opioid renewals require a written prescription, which offers an ideal opportunity to repeat depression screening using valid, brief assessments. Numerous options are available to screen for depression and mood disturbance in chronic pain [30,31,32,33]. The PHQ-2 [34], a two-item depression screener, is the shortest screening instrument and as accurate as the longer PHQ-9 [35] for detecting depression in patients with cancer [30]. Several measures, including the PHQ-9, Hopkins Symptom Checklist-20, and Mental Health Inventory-5 are all sensitive in detecting change in depression (improvement, worsening, or no change) among adult cancer patients with pain [31]. Many clinicians and researchers may prefer the PHQ-9 because it is available online and free to use.

While mental health screening in oncology is essential, screening alone is inadequate, and treating professionals need to be identified and engaged. Mental health specialists and psycho-oncologists are limited resources within cancer settings, though the ability to refer out to community psychiatry and psychology or engage in collaborative care models can alleviate this problem [36].

Treating depression in cancer pain

Guidance synthesizing evidence-based treatments for depression in the context of cancer pain is lacking. Practical management guidelines for depression in cancer do not address comorbid cancer pain [37•], and similar guidelines for cancer pain do not include treatment of depression [38]. However, stepped care approaches that guide management of both depression and pain in cancer can be extrapolated to treating depression in patients who also have cancer pain.

Psychosocial interventions are indicated for all classes of depression severity. Mindfulness and psychotherapy (broadly defined) both demonstrate benefit in depression and pain symptoms in patients with cancer, though studies are heterogeneous and evidence quality is low to moderate [39,40,41]. Cognitive behavior therapy, a mainstay in depression across other disease states, has only mixed evidence across cancer subtypes [41,42,43]. Beyond individual interventions, studies consistently demonstrate collaborative care as a superior model to usual treatment in cancer depression [44], with improvements in both depression and pain in the SMaRT Oncology-2 trial [45•]. While there are many collaborative care models for integrating primary care and mental health services, in the present case, collaborative care is a model that provides systematic, psychiatrist-guided depression interventions to a panel of patients, leveraging a limited pool of psychiatrists to provide evidence-driven treatment to more patients. Adapted from primary care to cancer settings, it partners a consulting psychiatrist with a care manager to review cases and engage with oncology medical teams to recommend a treatment plan and monitor response [45•].

Antidepressant medications, established as front-line therapy for depression in multiple disease states, have sparse and low-quality evidence for depression and pain across cancer settings. This includes a lack of benefit against placebo in a 2018 Cochrane review [46•] and inconclusive evidence (data only for amitriptyline and fluvoxamine) when added to opioids for cancer pain [47]. Thus, expert guidance recommends reserving antidepressants for moderate to severe depression in oncology settings [48••]. The SSRIs citalopram/escitalopram and sertraline are considered first-line, due to tolerability and low propensity for drug interactions [49]. SNRIs including duloxetine, venlafaxine/desvenlafaxine, and milnacipran/levomilnacipran, may be more beneficial in patients with comorbid neuropathic cancer pain or fibromyalgia [37, 48••]. The tricyclic antidepressants (imipramine, doxepin, nortriptyline, desipramine) are more poorly tolerated than SSRIs/SNRIs and pose higher risks in cancer patients given anticholinergic properties, cardiac toxicity, and cognitive impacts [50]. Nortriptyline and desipramine are less anticholinergic than others in this class.

Adjunctive, rapidly acting treatments for depression in cancer, such as psychostimulants (methylphenidate, dextroamphetamine) and stimulant-like medications (modafinil, armodafinil), bear consideration in patients with cancer pain given potential benefits for opioid-induced sedation [48••, 51]. Among these medications, data remain inconclusive but are strongest for methylphenidate in the treatment of depression generally [52] and in cancer settings [53].

Novel approaches to managing depression in cancer patients may also be relevant for patients with comorbid cancer pain, though evidence to guide practice is currently insufficient. Ketamine produced resolution of depression and suicidal ideation within 1 day in an RCT of depressed cancer patients [54] and has shown promise in cancer pain across case series and open-label studies. Unfortunately, ketamine failed to show reduced pain or opioid use in a meta-analysis of cancer pain [55], and a 2017 Cochrane review found low-quality, insufficient evidence for ketamine as an adjuvant to opioids in cancer pain [56]. In small clinical trials, psilocybin appears to ameliorate both depression and existential distress in cancer [57, 58]. To date, no published studies report on psilocybin’s impacts on comorbid depression and pain in patients with cancer.

Opioids and opioid use disorder in patients with cancer

Opioid-related mortality is 10 times lower in cancer survivors compared to the general population [59]. However, opioid-related deaths are increasing in cancer patients, which are thought to be related to increases in survival and subsequently higher levels of chronic pain and long-term opioid therapy [59]. With greater oversight on opioid prescribing, many clinicians have expressed concerns about undertreated cancer pain [60].

While opioids are the mainstay of treatment for cancer pain [61], oncologists do not know which patients they treat will become cancer survivors and whether they will develop non-medical opioid use behaviors [62]. Non-medical opioid use is the use of opioids without a prescription or in ways other than medically prescribed, including for the experience or feeling caused by the opioid [63]. The spectrum of non-medical opioid use may have features of varying degrees of opioid use disorder.

In a systematic review of 28 studies, Yusufov and colleagues [64••] found that substance use rates (variably reported, ranging from any substance use to a formal use disorder) ranged from 2 to 35% in cancer patients, with a median prescribed opioid rate of 18% across 7 studies reviewed. Survivors of cancer are also at risk of non-medical opioid use and opioid use disorder. In one study, the rate of opioid prescribing among cancer survivors was 1.22 times higher than among matched controls (95% CI, 1.11–1.3) [65]. Despite the potential risk and prevalence of opioid use disorder among patients with cancer and cancer survivors, there is a lack of confidence and training in providers managing opioid use disorder in patients with cancer [66, 67].

Compared to guidelines in patients with non-cancer pain, the best practices for identifying and providing optimal management of non-medical opioid use are less clearly defined for patients with cancer pain [68, 69••]. For example, existing CDC guidelines for safe opioid prescribing exclude patients with cancer-related pain who have distinct pain management needs that may require relatively high doses of opioids [69••, 70]. In addition, there is little outcomes-based research on how to manage non-medical opioid use in palliative care settings [64••].

Clinicians are encouraged to adopt universal precautions adapted from chronic non-cancer pain settings as their clinical framework for assessing and managing cancer-related pain [68, 71]. Adapted universal precautions include screening all patients for opioid misuse using brief tools, discussing the risks, benefits, adverse effects and alternatives of opioid therapy, and providing education on safe use, storage, and disposal [68, 72,73,74]. Screening for opioid use disorder should be done at initial consultation and regularly throughout treatment using a non-judgmental approach that allows patients the freedom to disclose if they are struggling with non-medical opioid use. Patients with risk factors for non-medical opioid use, such as a history of chronic non-cancer pain requiring opioids or comorbid psychiatric issues, should be referred to a specialist palliative care clinic or cancer pain clinic for co-management [69••, 74].

In a narrative review of non-medical opioid use, Ulker and Del Fabbro provide best practice recommendations on management of non-medical opioid use in patients with cancer-related pain [69••]. First, educational tools in the form of pamphlets or digital communication should be available for patients and their families about safe use, storage, and disposal of opioids. Strategies to mitigate the risk of non-medical opioid use and diversion include using pill counts and physician review of prescription drug monitoring systems that can provide information on whether the patient is receiving prescriptions from multiple providers or are being co-prescribed medications that increase the chance of overdose. The role of urine drug screens as a management strategy for reducing non-medical opioid use risk is unclear. There is weak evidence supporting the use of urine screens among patients with non-cancer pain and no current guidelines for patients with cancer [73].

Opioid agonists, including partial agonists like buprenorphine and full agonists like methadone, may have an important role in patients with cancer-related pain. Both medications are effective analgesics [75] and are also indicated treatments for opioid use disorder [76]. While a 2015 Cochrane review found that buprenorphine provides effective strong pain relief and offers an advantage of transdermal formulations, its role in the treatment of cancer pain is still evolving due to limited evidence [77]. In an updated Cochrane review, methadone was found to have similar benefits to morphine and was determined to have a role in the management of cancer pain based on low-quality evidence [78]. No systematic reviews have evaluated buprenorphine or methadone for treatment of opioid use disorder among patients with cancer. If patients require methadone or buprenorphine specifically for treatment of opioid use disorder, co-management with an addiction specialist may be indicated.

Use of adjuvant medications may reduce non-medical opioid use. However, research in this area is limited. A systematic review and meta-analysis of non-opioid analgesics in palliative medicine found evidence supporting pain reduction with the use of NSAIDs, but not with acetaminophen, either alone or in combination with strong opioids [79]. Another harm reduction strategy is limiting the use of sedatives, such as benzodiazepines, in combination with opioids, due to the increased risk of overdose death [80].

A key part of management of non-medical opioid use and opioid use disorder in cancer patients is managing psychological and spiritual distress, both of which can amplify expression of pain. In one small trial, patients with cancer pain who participated in an interdisciplinary intervention provided by a specialized team called the Compassionate High Alert Team (CHAT) had a significant reduction in the median (range) number of aberrant behaviors from 3 (1–6) pre-intervention to 0.4 (0–3) post-intervention (p<0.0001) [81•]. A modified version of the CHAT intervention has been implemented using telehealth visits to address non-medical opioid use among patients with cancer during the COVID-19 pandemic [82]. Case et al. [83] present the PARTNERS framework that deploys an interdisciplinary team of providers to support patients with cancer pain and opioid use disorder, using motivational interviewing to elicit and understand the patient’s perspectives and values to facilitate change. However, this approach has not been systematically studied. Individual clinicians who care for patients with cancer with non-medical opioid use should also be familiar with using approaches such as brief motivational interviewing, which is effective among certain populations but has not been evaluated in patients with cancer [69••, 84].

Conclusions

Beyond facing the direct challenges of cancer, oncology patients also bear increased risks of depression, pain, and the need for long-term opioids, which can predispose them to non-medical opioid use and opioid use disorder. The model of hyperkatifeia, while not yet studied in cancer, provides a theoretical framework for the role of long-term opioid therapy in depression and non-medical opioid use. Further research is needed as opioids remain a common approach to managing pain in cancer survivors, who now number about 17 million in the USA [85]. Understanding the interplay of these challenges, as well as their impacts on morbidity and mortality in cancer populations, empowers medical and psychosocial oncology providers to screen for, identify, and treat depression and opioid use disorder in patients with cancer pain. Recognizing a lack of clinical trials directly studying treatment of depression and opioid use disorder in cancer pain, we advocate for an increased research focus in this field.